Ankit Yadav

B.S. (Hons.) Applied Geology student · AI Engineering & Full-Stack Developer

Paschim Medinipur, India

@ankit4

0 followers

🎓 B.S. (Hons.) in Applied Geology at IIT Kharagpur · Graduating 2028

About

B.S. (Hons.) student at IIT Kharagpur with hands-on experience in AI engineering, retrieval-augmented generation, and full-stack web development. Built agentic browser automation and PDF question-answering systems using modern JavaScript/TypeScript, Python, and LLM tooling.

Experience

Tech Head

Ziksir

Developed frontend and backend features using Next.js, Express.js and TypeScript. Implemented OAuth-based authentication and authorization. Designed User, Booking and Order database schemas. Built secure cookie-based authentication and responsive UI components.

Education

2028

B.S. (Hons.)

IIT Kharagpur

Applied Geology · 2028

Applied Geology
2024

Senior Secondary Examination (Class XII)

Kendriya Vidyalaya Suratgarh Cantt

2024
2022

Secondary School Examination (Class X)

Kendriya Vidyalaya Suratgarh Cantt

2022

Skills

<3 months

C SQL CSS HTML React Python Bullmq Node.js Next.js LangChain Bitbucket LangGraph AI Agents OAuth 2.0 Express.js JavaScript TypeScript Authorization Authentication Prompt Engineering Application Development API Integration Understanding Retrieval Augmentation Generation

Tools / apps / platforms

<3 months

Git Codex Nginx Redis Cursor GitHub MongoDB PostgreSQL Claude Code Apache Kafka OpenAI ChatGPT

Projects

Retrieval-Augmented Generation (RAG) System for PDF Question Answering

LLMs, PDF parsing, embeddings, semantic retrieval

Developed a Retrieval-Augmented Generation (RAG) system for intelligent question answering over PDF documents. Implemented PDF parsing, text chunking, embedding generation, and semantic retrieval for accurate context extraction. Integrated Large Language Models (LLMs) with retrieved document context to generate context-aware responses. Optimized retrieval pipeline to improve answer relevance and reduce hallucinations. Built a modular architecture supporting multiple PDF documents and conversational document search.
Agentic Browser Harness for LLM Automation

Python, TypeScript, LangChain, Playwright

Built a Playwright-based browser harness enabling authenticated interaction with ChatGPT using a user's existing browser session. Reduced LLM testing costs to near-zero during development by routing requests through authenticated browser sessions instead of repeated API calls. Developed Python and TypeScript SDKs using LangChain to provide a unified interface for browser-based LLM interactions. Implemented tool-calling support, enabling the browser agent to interact with local applications and execute automation workflows. Enabled concurrent execution of multiple requests, improving testing throughput and developer productivity. Designed a modular SDK architecture supporting AI agents, browser automation, and extensible workflows.

Courses & certifications

Partial Differential Equations · IIT Kharagpur
Advanced Calculus · IIT Kharagpur
Linear Algebra · IIT Kharagpur
Probability and Statistics · IIT Kharagpur
Essentials of Machine Learning · IIT Kharagpur
Theory and Laboratory: Programming and Data Structures · IIT Kharagpur

⚽ Extracurricular activities

Tech Strategist, KGTS, IIT Kharagpur

Implemented leaderboard module, developed the Order and Chaos game feature, and contributed to technical planning and feature development.
Basketball

Secured third position at Shaurya Sports Event and third position in General Championship (GC) Basketball (202526).

Explore the Broxer community

💬 Q&A Ask about roles, salaries and interviews — and help others. 👥 Circles Join discussions on skills, companies and industries. 📰 Feed Follow updates and insights from the community.